Several communications systems employ synchronous operations, for which incoming and outgoing data flows, as well as data processing, are controlled by a timing clock. Multiplexing techniques, such as Time Division Multiplexing (TDM), enable the combination of several data channels onto a common channel by using prefixed time slots for each channel. Data reconstruction at the receiver should be synchronized to the multiplexing transmitter for reliable extraction of the desired information. Such synchronization strongly depends on a continuous and uninterrupted clock signal. The clock signal is usually originated at a stable and accurate oscillator, and is distributed to any required component by a plurality of CDUs. Therefore, a backup clock distribution circuit should be provided and held in standby mode, to function instead of the original circuit, whenever a failure is detected. Such failures may comprise a lost clock signal, reduction in its power level, frequency instabilities, changes of pulse width, etc.
A known method for providing a continuous clock signal to each desired point is to employ a pair of CDUs which consists of an active CDU and an additional redundant CDU. The output of each CDU is connected to a corresponding input of a selecting switch, which normally transfers the output of the active CDU to the output of the switch, connected to the desired circuitry or component which utilizes the clock signal. Whenever a failure is detected, the selecting switch selects the redundant CDU output and transfers it to its output. The switching operation between the two clock signals should be transparent to the circuitry (the load), fed by the clock signal. Therefore, the two clock signals should be continuously coherent, otherwise switching may cause phase discontinuities and undesired bit transitions. In addition, since the input (reference) signal should also be distributed to other circuits via auxiliary clock outputs, it is required to continuously keep the auxiliary clock signals of the active and redundant CDUs coherent, as well.
Phase-Locked-Loops (PLLs) are sometimes used after the selecting switch to smooth these transitions and to remove noise and instabilities from the clock signal. However, this solution is still problematic, since when using a single PLL and splitting the filtered clock signal, jitter and noise are accumulated along the clock signal path. This requires using such PLL circuitry in each input of a circuitry or component which utilizes the clock signal, and is therefore, costly. Moreover, even when using such plurality of PLLs, each of which smoothes incoming transients, during a long term, a phase difference will be developed, and perfect coincidence will not be obtained.
U.S. Pat. No. 4,282,493 discloses a clock signal generator for providing redundant clock signals, which comprises a master clock module and a slave clock module. The master and slave clock modules are always phase and frequency locked to one another. Upon detecting malfunction, the master is switched automatically or externally, between the clock modules. However, this construction is cumbersome, since each module comprises two PLL oscillators. In addition, this construction lacks the capability of separately controlling the response time of the locking circuitry, which causes smoother transition, and is not designed for the distribution of an externally fed clock signal, or of a higher frequency clock signal, based on a low frequency input signal.
U.S. Pat. No. 4,672,299 discloses a clock circuit which employs PLL circuits to synchronize between two input signals. During normal operation mode, the clock circuit is locked to one input signal. When switching to the other input signal, the PLL divider is controlled to force the loop phase to be matched to the phase of the newly selected input signal. However, this clock is not continuously synchronized to an input signal, but rather provides a phase correction in response to a phase error.
U.S. Pat. No. 5,422,915 discloses a fault tolerant CDU for providing synchronized clock signals to multiple circuit loads, which comprises oscillator circuitry, synchronization circuitry, selection circuitry and distribution circuitry, arranged in redundant form, so that partial failure will not result in total distribution failure. However, this CDU is complex and comprises a plurality of redundant sub-circuit. In addition, all redundant clock signals are locked to a reference signal, with no adaptation capability to be locked to each other.
U.S. Pat. No. 5,355,090 discloses a redundant clock system, generating active and standby clock signals in phase with one another through a pair of substantially identical cross-connected phase corrector circuits. Timing errors are reduced by causing all clock bus interface circuits to activate a standby corrector circuit and to cause the previously active standby corrector circuit to operate in standby mode. This system, however, requires two bus interface circuits and lacks adaptive operation of its employed PLLs to different requirements of response time.
All the methods described above have not yet provided satisfactory solutions to the problem of transparent switching between a Clock Distribution Unit (CDU) and a redundant CDU, while providing a redundant uninterrupted clock signal.
It is an object of the present invention to provide a method and apparatus for clock signal distribution with transparent switching between a Clock Distribution Unit (CDU) and a redundant CDU, which overcomes the drawbacks of prior art.
It is another object of the present invention to provide a method and apparatus for clock signal distribution, with transparent switching between a Clock Distribution Unit (CDU) and a redundant CDU, while keeping their corresponding clock signals continuously coherent.
It is still another object of the present invention to provide a method and apparatus for clock signal distribution with transparent switching between a Clock Distribution Unit (CDU) and a redundant CDU, with adaptive response time of their corresponding loop filters.
Other objects and advantages of the invention will become apparent as the description proceeds.